The Point Last Seen

Perhaps more than any other element, the “Point Last Seen” (PLS) is the most important part of an underwater recovery. This term is frequently used in Search And Rescue contexts, where information about where to start searching is critical to directing or driving a rescue or recovery. In underwater recoveries, searching an area of any size at all is frequently impossible. One diver might methodically cover an area of about 200’ x 100’ in one tank dive only if unimpeded by debris or rough underwater terrain. Most recovery dives are in far more complicated situations, significantly reducing the possible search area. This makes knowing exactly where the victim was last seen absolutely critical to the recovery process.

Obtaining an accurate PLS depends on the particular situation. Often, the best available information comes from a human witness. This is complicated by the trauma of witnessing a tragedy, along with a sort of “tunnel vision” that happens when an emergency is under way. Although difficult, the best approach with witnesses is to walk through the event on-scene, and ask about details that might help narrow down the search field. For example, we might ask an observer to place us at the exact location on the water of the event. We might also walk through mundane details, such as weather, time of day, who else was there, and so on, as this sometimes generates additional details. This can be a difficult balance, though, because sometimes people misremember if pressed too hard, especially if they’re trying to help. 

Also, witnesses are fallible, with distance estimates over water typically being dramatically short of actual distance due to lack of reference points and variable surface conditions. Often, our best source of information is electronic devices. Many boats, even rentals, have GPS equipped fishfinders. Obtaining the GPS track might indicate where a path started straying, if a fisherman went overboard, for example. It might also be associated with emergency phone calls, which would have a timestamp that could correlate to the GPS track. Another source that’s been helpful in the past: cell phone photographs. Even without location data, we’ve been able to narrow down the search area by lining up trees on the shoreline to match the photograph: height can record distance, while line-ups can record direction with decent accuracy. Also, a 911 call or EPIRB request optimally contains latitude and longitude, providing excellent PLS information.

Even without a fishfinder GPS track, or location metadata from a phone, an electronic footprint can be helpful. In several situations, we’ve been able to work with local authorities who provided access to cell phone tower signalling information. Cell phones provide a location by “triangulating” on three towers. Unfortunately, on water, this information can be rather coarse, providing locations that are around a quarter mile, unless the phone is fairly close to towers. Together with surface debris, currents, and a knowledge of how boats and vehicles sink, or how humans drown, this information can also be vital in narrowing a search start.

A few of the recoveries we’ve worked on over the past couple of years are illustrative. 

In one, an individual drowned in a deep lake from a rented boat. The depths were beyond a SCUBA search and recovery, but locating the individual using a side scan sonar towfish was also difficult: the underwater terrain was very steep and rocky, and, like many California reservoirs, dotted with oak trees, which can obscure items on the bottom, and easily snag a towfish and cable. After many difficult hours searching, we were able to speak with one of the witnesses, who provided very clear details, along with an exact time of making a 911 call. This allowed us to find a near exact location on the boat’s fishfinder/GPS unit, and an ROV recovery resulted. The combination of witness information with an electronic trail proved invaluable.

In another situation, a plane crashed in the Pacific Ocean. GPS information provided by the US Coast Guard, who lowered a rescue swimmer on the crash site as first responders, allowed us to start a sonar search of the area. Currents in the area swept crash debris into a nearby cove and out into the ocean, including the remains of one of the victims. Although the impact of the crash made a recovery impossible, we were able to search for an onboard flight recorder and other victims using towed sonar. Rough conditions in the cove prevented additional searching.

Another call we engaged in didn’t leave a GPS trail. However, a security camera happened to capture an image of our victim’s headlamp on a boat passing in front of a particular area in a bay popular for fishing. Tracking the path of that lamp, we were able to narrow down our search to a small area, and recover the victim within a day of searching.

Most recently, we’ve been helping with the recovery of two brothers who were duck hunting in an afterbay of Lake Oroville. We had a very specific location of a 911 call placed by one of the individuals before he attempted to rescue his brother from a capsized kayak. Unfortunately, the 911 call provided only a general location for where the brothers submerged. The kayak and other items were swept across the bay, around a half mile away. Unfortunately, the bay is filled with vegetation that grows over 6’ in height on the bottom. Like a drone flying over a thick forest, sonar scans are completely obscured by the thick layer of reeds/weeds. So hand searching was the only tool that can really work.

Methodical and systematic hand searches in weeds this thick are also extremely slow. Although the water is fairly shallow, a single diver might only be able to thoroughly search an area that's about 10’ x 30’ in a single dive. And maintaining a systematic record of searches over time, to know what has and has not been searched, presents additional challenges. Unfortunately, the boys were not located, even though many law enforcement dive teams, sonar and ROV teams, and groups of private and commercial divers have searched. After a couple of days searching with sonar, our team spent another day attempting to clear the highest probability regions, and decided to stand down.

This is a difficult decision to make. The shallow water (11-20’ deep) means nature did return the boys eventually: when this was first written, one had been recovered, while the family waited for the other. But the extremely slow hand searching would likely take more time than nature, so the right decision was to do what we can, and then let nature take its course. Just as this post was published, nature returned the second brother to the surface.

An accurate Point Last Seen is our most important tool. When we don’t have it, we can sometimes rely on technology, knowledge of currents, or our own intuition and experience coupled with  technology. But when the conditions don’t allow us to search a large area quickly, and we don’t know exactly where to focus despite applying a lot of thought, we stand down and respect what nature gives us.